| Summary: | 碩士 === 國立東華大學 === 公司理財碩士學位學程 === 97 === The capital asset pricing model (CAPM) is the important cornerstone for modern financial theory. However, CAPM is constantly challenged by other asset pricing models, and Fama-French three-factor model is the most famous among those models.
This paper analyzes asset pricing models using the nonparametric regression method to estimate stochastic discount factors. To avoid the problems of data mining and functional form misspecification, the capital asset pricing model (CAPM; Sharpe, 1964; Lintner, 1965), the three-factor model (Fama and French, 1993), the four-factor model (Carhart, 1997) and the five-factor model are improved using a nonparametric regression method with conditional information. Here, the five-factor model considers those factors used in the four-factor model and illiquidity factor (Amihud, 2002). Based on the data collected from the Tokyo stock exchange in Japan, the conditional five-factor model yields the minimum pricing error. Under the significance level 0.05, the adequacy of the model is supported. Consequently, the conditional five-factor model is far better than other models of the Japanese stock market.
|
|---|
